ORIGINAL ARTICLE Chronic lymphocytic leukemia in Korean patients: frequent atypical immunophenotype and relatively aggressive clinical behavior Mi-Ae Jang • Eun-Hyung Yoo • Kihyun Kim • Won Seog Kim • Chul Won Jung • Sun-Hee Kim • Hee-Jin Kim Received: 20 August 2012 / Revised: 23 January 2013 / Accepted: 23 January 2013 / Published online: 12 February 2013 � The Japanese Society of Hematology 2013 Abstract Chronic lymphocytic leukemia (CLL) is a mature B-cell neoplasm characterized by the expansion of CD5-positive lymphocytes in peripheral blood. While CLL is the most common type of leukemia in Western popula- tions, the disease is rare in Asians. Hence, clinical and laboratory data and studies of CLL in Asian populations have been limited. In this study, we investigated the clin- ical and laboratory characteristics of CLL in Korea. A total of 39 patients who had been diagnosed with CLL during the period from January 2000 to October 2010 at a single institution in Korea were examined. Clinically, 67 % of the patients were classified as having advanced Binet stages B or C. Up to 56 % of the patients had an atypical immu- nophenotype with high frequencies of FMC7 positivity and strong CD22 positivity. Twenty-six patients (67 %) received chemotherapy, and more than half of the treated patients (54 %) expired. The overall survival rate at 5 years was estimated at 71 %, which was lower than previously reported. These findings suggested that CLL in Korea has atypical immunophenotypes and that its clinical behavior may be more aggressive than that in Western populations. Keywords Chronic lymphocytic leukemia � Atypical � Immunophenotype � Korea Introduction Chronic lymphocytic leukemia (CLL) is a mature B-cell neoplasm characterized by the expansion of CD5-positive small- to medium-sized lymphocytes in peripheral blood and clinical manifestations of lymphadenopathy, spleno- megaly, and hepatomegaly [1]. The laboratory criteria for the diagnosis of CLL are well established and require the presence of at least 5 9 109 B lymphocytes/L in peripheral blood and the confirmation of the clonality of the B cells by flow cytometry [2]. The CLL cells immunophenotypically coexpress CD5 and CD23 and have a restrict pattern of surface immunoglobulin expression, kappa or lambda [2]. Men are more commonly affected than women, with a median age of diagnosis of approximately 65–70 years. CLL typically has an indolent clinical course; however, a subset of patients with CLL present with a more aggressive phenotype and worse prognosis [3]. According to the pre- vious studies, atypical morphologic features and cytoge- netic abnormalities such as trisomy 12 seem to be markers of aggressive clinical behavior in CLL [3–5]. Although CLL is the most common leukemia in western countries, the prevalence of CLL in Asian populations is very low. The incidence rate of CLL is about 2–6 cases per 100,000 people per year in the USA, increasing with age to 12.8 per 100,000 people per year at age 65, which is the mean age at diagnosis [6]. By contrast, the incidence rate of CLL in Japan is 0.48 per 100,000 people per year, M.-A. Jang � E.-H. Yoo � S.-H. Kim (&) � H.-J. Kim (&) Department of Laboratory Medicine and Genetics, Samsung Medical Center, Sungkyunkwan University School of Medicine, 81 Irwon-ro, Gangnam-gu, Seoul 135-710, Korea e-mail:
[email protected] H.-J. Kim e-mail:
[email protected] Present Address: E.-H. Yoo Department of Laboratory Medicine, Konyang University Hospital, Konyang University College of Medicine, Daejon, Korea K. Kim � W. S. Kim � C. W. Jung Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea 123 Int J Hematol (2013) 97:403–408 DOI 10.1007/s12185-013-1286-z accounting for only 3 % of all leukemia cases [7]. The incidence of CLL in Korea shows a tendency to increase under the influence of an aging society, westernized diets, and environmental factors than in the past [8]. However, a recent research from Korea still showed the low incidence rate of lymphoid leukemia including CLL, only 1.2 per 100,000 in Korea [9]. Accordingly, most of the studies on CLL have been reported from western countries, and data from Asian populations have been limited. In addition to the striking difference in the incidence, a recent study from Japan reported that up to half (48 %) of CLL cases had atypical immunophenotypes [10]. In the present study, we investigated the clinical and laboratory findings of Korean CLL patients and compared them with those of western and Japanese CLL patients. Patients and methods Patients The study patients were those diagnosed with CLL at Samsung Medical Center between January 2000 and October 2010. The diagnosis was made based on mor- phologic and immunophenotypic (immunohistochemical and flow cytometric) findings according to the World Health Organization classification [1]. To exclude mantle cell lymphoma, we confirmed the lack of t(11:14)(q13;q32) by fluorescence in situ hybridization (FISH) or negative cyclin D1 immunostaining. The translocation t(14;18), which involve the BCL2 genetic hallmark for follicular lymphoma, was also excluded from the analyses. All of the patients were classified by the Binet staging system as one of three subgroups (stage A, B and C) based on the number of involved areas (defined by the presence of enlarged lymph nodes[1 cm in diameter or organomegaly) and the presence of anemia (\10 g/dL) or thrombocytopenia (\100 9 109/L). Clinical information included age, gen- der, presence or absence of peripheral lymphadenopathy, splenomegaly, hepatomegaly, and extranodal involvement, treatment history and the cause of death. As for the labo- ratory data, total white blood cell counts, absolute lym- phocyte counts, hemoglobin level, and platelet counts at diagnosis were reviewed along with the calculation of lymphocyte-doubling time (LDT) as a prognostic factor when the data were available [2]. This study was approved by the Institutional Review Board of Samsung Medical Center. Immunophenotypic analyses In each patient, immunophenotyping of leukemic cells was performed by flow cytometry using the dual-laser FACSCantoTM II flow cytometer (Becton–Dickinson, San Jose, CA, USA) with a panel of lymphoid cell-associated monoclonal antibodies for CD2, CD3, CD5, CD7, CD10, CD19, CD20, CD22, CD23, CD25, FMC7, nuclear TdT, and surface immunoglobulin (Becton–Dickinson). Data were acquired and analyzed with the BD FACSDivaTM software (Becton–Dickinson). The immunophenotype score was calculated based on the immunophenotypic CLL scoring system proposed by Matutes et al. [11]. Cytomorphology The cytomorpholgic diagnosis was made based on the findings from forced air-dried, Wright-Giemsa-stained blood film preparations as a routine diagnostic procedure. Typical CLL were small- to medium-sized monomorphic lymphocytes with clumped chromatin and scanty cyto- plasm. The larger cells with prominent nucleoli were counted as prolymphocytes, which usually comprised\2 % of the peripheral blood smear. If the prolymphocyte popu- lation constituted more than 10 % and fewer than 55 % of the circulating lymphocytes in peripheral blood, the diag- nosis of CLL/prolymphocytic leukemia (CLL/PL), a mor- phologic variant of CLL, was made. Cytogenetics and fluorescence in situ hybridization Conventional cytogenetic studies were performed on hep- arinized bone marrow samples of the study patients. Each sample was cultured for 24 h and 72-h of lipopolysaccha- ride stimulation using the protocol for routine clinical cancer cytogenetics laboratory. After harvest, cells were treated with hypotonic solution, fixed in methanol/acetic acid, and G banded according to the standard methods (3:1 ratio). At least 20 metaphases were analyzed for karyo- typing. To rule out the translocation involving IGH and CCND1 genes, FISH study was performed on interphase nuclei using the LSI IGH/CCND1 dual-color, dual-fusion translocation probe (Vysis Inc., Downers Grove, IL, USA) according to the manufacturer’s instructions. Statistical analyses and survival analyses Statistical analyses were performed using a Chi-square test and Fisher’s exact test for categorical variables and Mann– Whitney U test or one-way analysis of variance for con- tinuous variables, as appropriate. The survival analyses were performed by the Kaplan–Meier plots. The overall survival (OS) was determined from the time of initial diagnosis to death or final follow-up in total patients and according to the tumor characteristics (immunophenotype scores, cytomorphology, and the clinical stage). The dif- ferences in OS were compared for a statistical significance 404 M.-A. Jang et al. 123 using the log-rank test. The comparison of 5-year OS in patient subgroups was performed using the Z score. A difference with a P value\0.05 was considered statistically significant. All statistical analyses were performed using the statistical Software Package for the Social Sciences (IBM SPSS Statistics 19; SPSS Inc., Chicago, IL, USA). Results Patient characteristics and laboratory findings Fifty-three patients were diagnosed with CLL during the study period. Among them, 14 patients were excluded due to incomplete clinical data, and a total of 39 patients were included for analyses. The characteristics of the 39 patients are shown in Table 1. Among the patients, 23 were male and 16 were female (a male-to-female ratio of 1.4:1), with a median age of 61 years (range 32–93 years). The median hemoglobin concentration was 13.1 g/dl, and the median platelet count was 161.5 9 109/L. The median white blood cell count was 16.1 9 109/L, and the absolute lymphocyte count was 8.4 9 109/L. Four patients (10 %) had anemia (hemoglobin \10 g/dL). Seven patients (18 %) had thrombocytopenia (platelet count \100 9 109/L). The clinical stage of the disease was as follows: Binet stage A in 13 patients (33 %), stage B in 16 patients (41 %), and stage C in 10 patients (26 %). LDT was calculated in 19 patients. Eleven patients had an LDT \6 months (58 %), and 82 % of them (9/11) had advanced Binet stage (B in 7 patients and C in 2 patients). Immunophenotypic findings The immunophenotypes of the 39 study patients are shown in Table 2. CD19 and CD5 were positive in all cases (100 %), CD23 was positive in 38/39 cases (97 %). When the immunophenotypic CLL scoring system by Matutes et al. was applied, 17 patients (44 %) had the high-score CLL ([3 points), and 22 patients (56 %) had the low-score CLL (B3 points). The positivities for FMC7 (49 %, 19/39) and CD22 (86 %, 25/29) were significantly different between the current study and the previous study by Matutes et al. (FMC7, 49 vs. 19 %, P \ 0.001; CD22, 86 vs. 7 %, P \ 0.001). CD23 was negative in one case (3 %). This case also had a strong reactivity for CD22, which is an atypical finding for CLL. A strong expression of CD20 was observed in 22/39 cases (56 %). Cytomorphologic findings In 31 out of 39 cases (79 %), the CLL cells had typical features, that is, small- to medium-sized lymphocytes with Table 1 Characteristics of 39 Korean patients with chronic lym- phocytic leukemia Characteristics Median age at diagnosis, years (range) 61 (32–93) Sex Male:female 1.4:1 Binet stage A 13 (33 %) B 16 (41 %) C 10 (26 %) Cytogenetics Normal karyotype 24 (69 %) Abnormal karyotype 11 (31 %) Trisomy 12 6 11q deletion 2 Complex karyotype 2 Unbalanced 1;7 translocation 1 No mitotic cells 1 Not available 3 Chemotherapy Not done (all alive) 13 (33 %) Done 26 (67 %) Died 14 Alive 11 Follow-up loss 1 Table 2 Immunophenotypic features in 39 Korean patients of chronic lymphocytic leukemia Marker Current study Western studya P valueb % Positive/ tested % Positive/ tested Surface immunoglobulin Kappa 49 19/39 56 228/396 0.369 Lambda 46 18/39 34 136/396 Negative 5 2/39 8 32/396 Intensity Weak (±) 73 27/37 75 298/400 0.810 Moderate(?) 24 9/37 23 93/400 Strong (??) 3 1/37 2 9/400 CD5 100 39/39 92 366/398 0.098 CD23 97 38/39 94 352/376 0.494 FMC7 49 19/39 19 69/386 \0.001 CD22 Weak (±) 14 4/29 40 140/355 \0.001 ?/?? 86 25/29 7 26/355 a Matutes et al. [11] b P values are from the comparison of the current and previously published studies Atypical characteristics of CLL in Korea 405 123 clumped chromatin and scanty cytoplasm. Eight cases (21 %, 8/39) had CLL/PL. Morphologic variants other than CLL/PL, such as lymphoplasmacytic or cleaved cells, were not observed in our series. Clinically, the atypical CLL/PL group had higher white blood cell counts (median 27.0 vs. 11.4 9 109/L, P = 0.095), higher absolute lymphocyte count (median 23.6 vs. 7.1 9 109/L, P = 0.067) and lower platelet count (median 113 vs. 171 9 109/L, P = 0.047) at diagnosis than in the typical morphologic group, although these differences were not significant statistically. None of the patients with the CLL/PL group presented with Binet stage A (total 8 patients: B stage 4 and C stage 4). The proportion of treated patients was also higher in the CLL/ PL group (88 %, 7/8) than in the typical morphologic group (61 %, 19/31), but the difference was not significant (P = 0.394). The abnormal karyotype was significantly more frequent in the CLL/PL group than in the typical CLL group (75 %, 6/8 vs. 19 %, 5/27, P = 0.006). Four patients in the CLL/PL group had trisomy 12 (50 %, 4/8). Cytogenetics and FISH Of the 39 patients with CLL, the cytogenetic study was not successful in 3 patients due to the poor quality of specimen. In 1 patient, cytogenetic study yielded no mitotic cells. Abnormal cytogenetic findings were noted in 11 out of 39 cases (28 %), including trisomy 12 (n = 6), 11q deletion (n = 2), complex karyotype (n = 2) and unbalanced 1;7 translocation (n = 1). Among trisomy 12 cases, 2 cases were associated with translocation involving the immuno- globulin heavy chain gene (IGH, 14q32 locus) with t(14;19)(q32;q13.3). One patient showed a derivative chromosome 7 derived from a translocation between 1q21 and 7q34, resulting in 1q21-1qter trisomy and 7q34-7qter monosomy. The FISH analyses for IGH/CCND1 demon- strated that no patient in our series had the rearrangement. Treatment and survival The median time of follow-up for the 39 study patients was 41 months (range 1–192 months). Twenty-six patients (67 %) required treatment during the follow-up period. The initial treatment included chlorambucil or fludarabine. Six patients received rituximab in addition to fludarabine. Of the treated patients, more than half (14/26, 54 %) were died. Of 13 patients with Binet stage A, 4 (31 %) patients received chemotherapy: 2 patients received combination chemotherapy on the initial diagnosis due to the involve- ment of the breast and stomach, respectively, and 2 patients was initiated with chlorambucil monotherapy at 6 and 21 months from the diagnosis. Nine out of 11 patients (82 %) with LDT\6 months, excluding 2 with Binet stage A, received treatment. The OS rate at 5 years in the total patients was estimated at 71 %, with a median OS of 93 months (95 % confidence interval, 55–131 months (Fig. 1a). According to the immunophenotype, the OS was not superior in the patients with score[3 (typical) vs. those with score 3 or lower (atypical), but without a statistical significance (P = 0.775, Fig. 1b). According to the cyto- morphology, the OS was superior in those with the typical CLL vs. those with CLL/PL, but without a statistical sig- nificance (P = 0.058, Fig. 1c). According to the Binet stage, the OS was highest in those with stage A, followed by stage B and C, but without a statistical significance (P = 0.063, Fig. 1d). Among the 14 patients (36 %, 14/39) who expired, 1 patient with Binet stage C died after a period of only 1 month from diagnosis. The 5-year OS rates for cytomorphology had statistical significance (P = 0.003), but no other comparisons had statistical significance. Discussion Due to the low incidence of disease, data on CLL have been limited in Asian countries including Korea. The current study involves the largest number of CLL patients in Asian countries. In comparison to the immunopheno- typically typical and clinically indolent manifestations of Western CLL, we found that Korean patients with CLL frequently manifest with atypical immunophenotype and had a more aggressive clinical behavior. First, we found that Korean CLL patients were more likely to have atypical immunophenotypes. In 1994, Matutes et al. proposed an international scoring system based on the expression of 6 surface antigens to facilitate the differential diagnosis between CLL and other mature B cell neoplasms. A higher score ([3) means the typical immunophenotype, and a low score (B3) means atypical immunophenotype. Based on these criteria, the majority of CLL cases (87 %) had high scores (4 and 5), while only a minority (3 %) had low scores [11]. By contrast, more than half of the Korean CLL patients (22/39, 56 %) in our series had low scores (B3 points), that is, atypical immunophe- notype. Although the FMC7 negativity has been known to be one of the most reliable markers to distinguish CLL from other B-cell neoplasms [11], we observed a high proportion of FMC7 positivity (49 %) in our series. Atypical features also include strong expression of CD22 (86 %) and CD20 (56 %). Of interest, the study by Tomomatsu et al. [10] from Japan also reported on the high proportion of low-score CLL (B3 points) in Japanese patients with CLL (12/28, 48 %) and strong reactivity of CD20 (20/28, 71 %). Along with this report from Japan, our data suggest frequent atypical immunophenotypes of CLL in Asia. Cytomorphologically, 21 % of CLL in our 406 M.-A. Jang et al. 123 series had the CLL/PL variant. The CLL/PL variant was proposed by the French–American–British group and is defined as the presence of a dimorphic population of small lymphocytes and prolymphocytes in the peripheral blood, having the prolymphocyte population 10–55 % of circu- lating lymphocytes. Prolymphocytes [55 % favor the diagnosis of B-cell prolymphocytic leukemia according to the WHO classification [1]. There have been a number of studies documenting the association between CLL/PL and the presence of trisomy 12, advanced clinical stage, and a higher proliferative rate. In line with these previous observations, the CLL/PL cases in our series had higher WBC counts, higher lymphocyte counts and lower platelet counts at diagnosis, compared to the typical CLL patients. The proportions of advanced Binet stages and cases requiring treatment were significantly more frequent in the CLL/PL group. Adverse prognostic markers on cytoge- netics, such as trisomy 12, were also more frequently observed in the CLL/PL group. Clinically, about two-thirds of the study patients (67 %) had advanced Binet stages (B or C). Compared with Western and Japanese CLL patients, this proportion was relatively high (40 % in Western and 36 % in Japan). The Binet staging is a widely used system to assess prognosis, having 3 groups with different prognosis. The majority of CLL patients has been known to have Binet stage A, the lowest risk group. In general practice, patients with Binet stage A should be monitored without therapy. On the other hand, Binet stages B or C represent the high risk group, comprising up to 30 and 10 % of the Western CLL patients, respectively [12]. These advanced groups usually benefit from the initiation of treatment [2]. The higher proportion of patients with advanced Binet stage in the present series was considered to be associated with a high proportion of patients with short LDT [82 % of patients with LDT \6 months (9/11) had advanced Binet stages (B or C)]. In our series, 67 % of our patients required chemother- apy, and 54 % of those who received treatment expired. The overall survival rate at 5 years in 39 patients was estimated at 71 %, and the estimated median overall survival was only 93 months (95 % confidence interval 55–131). These results are different from common manifestations of highly stable, smoldering, low-risk disease in CLL cases in Wes- tern and Japanese populations [10, 12–14]. However, it is far too early to generalize the aggressive clinical behavior based upon data from a single institution. Since this was a retrospective study, further investigation Fig. 1 a Kaplan–Meier plots for the overall survival (OS) estimation. The OS rate at 5 years in 39 study patients was estimated at 71 %, and the median survival was 93 months (95 % confidence interval 55–131 months). b OS according to the immunophenotype (IP): typical CLL (score [3; N = 17) group vs. atypical CLL (score C3; N = 22) group (log-rank P = 0.775). c OS according to the cytomorphology: typical CLL (N = 31) vs. CLL/ prolymphocytic leukemia (CLL/ PL) group (N = 8) (log-rank P = 0.058). d OS according to Binet stage: Binet A (N = 13), Binet B (N = 16), and Binet C (N = 10) (log-rank P = 0.063) Atypical characteristics of CLL in Korea 407 123 for CLL prognostic markers, such as b2-microglobulin level, expression CD38 and ZAP-70, was inadequate. In addition, selection bias could be possible. Our institution is a tertiary care hospital, in which the adverse clinical out- come from the patient with advanced stage could be overestimated. Further accumulation of data from multi- institutional studies could be helpful to overcome these pitfalls. The correlation of immunophenotype, cytomorphology, and clinical behavior in our series of Korean patients with CLL is illustrated in Fig. 2. Although the previous study documented that patients with the typical morphology tend to have higher immunophenotype scores than those with an atypical morphology, this was not apparent in this study. In our series, a significant proportion of patients in the typical morphology group had lower immunopheno- type scores. Also, the atypical morphology group, CLL/PL was equally distributed into high or lower score group of immunophenotype. In conclusion, the present study demonstrated clinical and laboratory characteristics of CLL in Korea. The observation of the frequent presentation with an advanced clinical stage and atypical immunophenotype and cyto- morphology further expands the heterogeneity of the dis- ease, particularly in terms of ethnic differences. References 1. Swerdlow SH, Campo E, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman JW. WHO classification of tumours of hae- matopoietic and lymphoid tissues. 4th ed. Lyon: IARC; 2008. 2. Hallek M, Cheson BD, Catovsky D, Caligaris-Cappio F, Dighiero G, Dohner H, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood. 2008;111:5446–56. 3. Frater JL, McCarron KF, Hammel JP, Shapiro JL, Miller ML, Tubbs RR, et al. Typical and atypical chronic lymphocytic leu- kemia differ clinically and immunophenotypically. Am J Clin Pathol. 2001;116:655–64. 4. Matutes E, Oscier D, Garcia-Marco J, Ellis J, Copplestone A, Gillingham R, et al. Trisomy 12 defines a group of CLL with atypical morphology: correlation between cytogenetic, clinical and laboratory features in 544 patients. Br J Haematol. 1996; 92:382–8. 5. Matutes E, Polliack A. Morphological and immunophenotypic features of chronic lymphocytic leukemia. Rev Clin Exp Hema- tol. 2000;4:22–47. 6. Rozman C, Montserrat E. Chronic lymphocytic leukemia. N Engl J Med. 1995;333:1052–7. 7. Tamura K, Sawada H, Izumi Y, Fukuda T, Utsunomiya A, Ikeda S, et al. Chronic lymphocytic leukemia (CLL) is rare, but the proportion of T-CLL is high in Japan. Eur J Haematol. 2001; 67:152–7. 8. Lee JJ, Ahn JS. Recently increasing hematologic diseases in Korea. Korean J Med. 2010;78:557–63. 9. Park HJ, Park EH, Jung KW, Kong HJ, Won YJ, Lee JY, et al. Statistics of hematologic malignancies in Korea: incidence, prevalence and survival rates from 1999 to 2008. Korean J Hematol. 2012;47:28–38. doi:10.5045/kjh.2012.47.1.28. 10. Tomomatsu J, Isobe Y, Oshimi K, Tabe Y, Ishii K, Noguchi M, et al. Chronic lymphocytic leukemia in a Japanese population: varied immunophenotypic profile, distinctive usage of frequently mutated IGH gene, and indolent clinical behavior. Leuk Lym- phoma. 2010;51:2230–9. 11. Matutes E, Owusu-Ankomah K, Morilla R, Garcia Marco J, Houlihan A, Que TH, et al. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia. 1994;8:1640–5. 12. Oscier D, Fegan C, Hillmen P, Illidge T, Johnson S, Maguire P, et al. Guidelines on the diagnosis and management of chronic lymphocytic leukaemia. Br J Haematol. 2004;125:294–317. 13. Guarini A, Gaidano G, Mauro FR, Capello D, Mancini F, De Propris MS, et al. Chronic lymphocytic leukemia patients with highly stable and indolent disease show distinctive phenotypic and genotypic features. Blood. 2003;102:1035–41. 14. Wierda WG, O’Brien S, Wang X, Faderl S, Ferrajoli A, Do KA, et al. Prognostic nomogram and index for overall survival in previously untreated patients with chronic lymphocytic leukemia. Blood. 2007;109:4679–85. Fig. 2 A graphical illustration on the relationship between immuno- phenotype, cytomorphology, and clinical behavior in study patients. Numbers in parentheses represent the number of patients. Top immunophenotype; high-score group [3 points; low-score group B3 points. Middle cytomorphology; the typical CLL type and the CLL/ prolymphocytic leukemia (CLL/PL) variant type. Bottom clinical behavior; Binet stage A, B, and C 408 M.-A. Jang et al. 123 Chronic lymphocytic leukemia in Korean patients: frequent atypical immunophenotype and relatively aggressive clinical behavior Abstract Introduction Patients and methods Patients Immunophenotypic analyses Cytomorphology Cytogenetics and fluorescence in situ hybridization Statistical analyses and survival analyses Results Patient characteristics and laboratory findings Immunophenotypic findings Cytomorphologic findings Cytogenetics and FISH Treatment and survival Discussion References